Yuelyu J
Pittsburgh, Pennsylvania, United States
Summary
👤
Senior🎓
Top SchoolYuelyu J is an applied scientist and Information Science PhD candidate at the University of Pittsburgh with eight years of experience building and evaluating retrieval-augmented generation (RAG), multimodal reasoning, and trustworthy AI systems. Her work—published at SIGIR, ACL, and npj Digital Medicine—focuses on reasoning-based knowledge distillation, fairness-aware training, and bias mitigation for clinical AI, and she has translated that research into production experiments at NetEase and AWS to improve user satisfaction through RL-enhanced text generation. At NetEase she helped create the COCO-Music dataset and a Continuous Parameterization approach that balances factuality and emotional style using PPO, a less obvious blend of music-domain data creation and reinforcement learning. Now at Microsoft, she continues to bridge rigorous academic methods with applied engineering to build reliable, equitable multimodal AI, and she is actively seeking collaborations in RAG, multimodal systems, and responsible AI.
8 years of coding experience
2 years of employment as a software developer
南京农业大学Doctor of Philosophy - PhD, Information Science, Doctor of Philosophy - PhD, Information Science at University of Pittsburgh
Github Skills (16)
alpaca6
huggingface4
lora4
transformers4
pytorch4
few-shot-learning4
backend4
django3
chatgpt3
language-model3
fine-tuning2
rlhf2
peft2
prompt2
sequence-labeling2
Programming languages (3)
Github contributions (5)
JoyDajunSpaceCraft/Leetcode_2024_summer
May 2024 - Sep 2024
Contributions:127 pushes, 1 branch in 4 months
JoyDajunSpaceCraft/leetcode_job
Feb 2020 - Jun 2020
Contributions:71 pushes, 1 branch in 4 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.